Every change Wellknown observed on this MCP server, newest first, with what it was before and what it became. Tool-surface changes carry the definition diff. Nothing here is edited after the fact.
Certificate recorded, valid to 2026-12-19
Authorization not required
Added "get_balance"; changed the definition of "chat_text", "generate_image", "get_job" and 1 more (11 tools before, 12 now)
Private LLM text completion (paid, x402)
Get the paid endpoint forOpenAI-compatible chat completions — frontier and open-weight models (Claude, DeepSeek, GLM, GPT, Gemini, Grok, Kimi, Llama, Qwen;list under textModels in GET /v1/agent/models).PAIDPAID.call,Iftwothisrails:MCPx402request carries your API key (endpointx-api-keyanswers/402,Bearerwalletlp_…)
Generate an image or video (paid, x402)
Render or transform an image or video with a catalog model (t2i, i2i, t2v, i2v) — including the private model line flagged private=true in list_models. This is a PAID call,two rails: (1) x402tool —thetoolsendpointlikeanswersupscale/remove-backgroundHTTPtake402image withpaymentnorequirementsprompt)
Check a running job returned bythe paid generate endpointgenerate_image — free. Returns {status:'done',url} | running | failed.
⟨6 unchanged words⟩ "jobId":{"description":"jobId returned by/v1/agent/generatethe generate endpoint","type":"string"}},"required": ⟨3 unchanged words⟩
{"destructiveHint":false,"idempotentHint":true,"openWorldHint":false,"readOnlyHint":truefalse}
⟨14 unchanged words⟩ Cheapest, fastest, or highest quality; for images, video, orvideo;utility tools (upscale, background removal); with or without a source image. Pass private ⟨12 unchanged words⟩
⟨22 unchanged words⟩ a source image drives the job (forcesi2i/i2vi2i/i2v/tool)","type":"boolean"},"media":{"description":"Output type — 'tool' = image transforms (upscale, remove-bg)","enum":["image","video","tool"],"type":"string"},"private": ⟨17 unchanged words⟩
Check prepaid credit balance — Prepaid credit balance and recent ledger for the API key on this MCP request — needs an lp_ key (x-api-key / Bearer).
Changed the definition of "chat_text" and "generate_image"
⟨6 unchanged words⟩ chat completions — frontier and open-weight models (DeepSeekClaude,GPTDeepSeek, GLM,QwenGPT, Gemini,LlamaGrok, Kimi,GrokLlama,Claude;Qwen; list under textModels in GET /v1/agent/models). PAID ⟨18 unchanged words⟩ key (`x-api-key: lp_…` — create at https://livepairai.com/dashboard/settings//livepairai.com/settings after signing up; bills prepaid credits, no ⟨27 unchanged words⟩
Added "chat_text"; changed the definition of "generate_image" (10 tools before, 11 now)
⟨20 unchanged words⟩ private=true in list_models. This is a PAID call, two rails: (1) x402 — the endpoint answers HTTP 402 withx402payment requirements (USDC on Base mainnet — priced per call, video billed per second per model rate). Retry the;requestretry with a PAYMENT-SIGNATURE header settling one requirement—(use an x402-capable client(@x402/fetch orlikeequivalent@x402/fetch). (2) API key — sign up at https://livepairai.com/auth/sign-up, create a key at /dashboard/settings, send `x-api-key: lp_…`; the call bills your prepaid credit balance and skips the 402 entirely.
Changed the definition of "enhance_prompt", "generate_image", "list_models" and 3 more
Rewrite a rough prompt idea into the structured formLivePair'sour image models follow best (subject, composition, ⟨11 unchanged words⟩
Render an image or video with aLivePaircatalog model (t2i, i2i, t2v, i2v).— including the private model line flagged private=true in list_models. This is a PAID call: the endpoint ⟨58 unchanged words⟩
Added "enhance_prompt", "generate_image", "get_job" and 3 more (4 tools before, 10 now)
Improve a generation prompt — Rewrite a rough prompt idea into the structured form LivePair's image models follow best (subject, composition, style, negatives). Use before generate_image — better input, better output.
Generate an image or video (paid, x402) — Render an image or video with a LivePair model (t2i, i2i, t2v, i2v). This is a PAID call: the endpoint answers HTTP 402 with x402 payment requirements (USDC on Base mainnet — priced per call, video billed per second per model rate). Retry the request with a PAYMENT-SIGNATURE header settling one requirement — use an x402-capable client (@x402/fetch or equivalent). On success returns {status:'done',url}; slow jobs return a jobId to poll for free.
Poll a generation job — Check a running job returned by the paid generate endpoint — free. Returns {status:'done',url} | running | failed.
Prompt pack by topic — A themed bundle of curated prompt picks (themes: anime, portrait, landscape, product, character, sci-fi, fantasy, abstract, food, architecture). Curated packs are rebuilt daily and carry real paid-run counts — prompts that already produced paid results. Falls back to a live topic search when no pack covers the theme.
Changed the definition of "trending_prompts"
⟨17 unchanged words⟩ for these. Returns community prompt cards ranked byviews, likes,thecommentspaidandgenerations'trytheyit'actuallyclicks.drove.
Unknown → Live
First tool surface recorded: 4 tools (server version 0.1.0)
https://livepairai.com/mcp (mcp_streamable_http) — from mcp_registry, with the record
Showing the latest 11 events. The API returns up to 500 and filters by kind: ?kind=tool_surface_changed
⟨7 unchanged words⟩ description":"Output cap, billed at this amount(<= 8192, default 1024)","type":"number"},"model":{"description":"Text model id (default:seedeepseek/deepseek-v4.1-flashtextModels in list_models)","type":"string"},"payer":{"description":"Your 0x… wallet —applies yourvolume-tierdiscount; must equal the wallet signingdiscountthe(x402paymentrail)","type":"string"},"prompt":{ ⟨14 unchanged words⟩
{"destructiveHint":false,"idempotentHint":false,"openWorldHint":truefalse,"readOnlyHint":false}
⟨16 unchanged words⟩ },"image":{"description":"Source imagefor—i2i/i2vrequiredmodelsfor—i2i/i2v/toolpublicmodels. Public http(s) URL or data:image URI. Use ⟨33 unchanged words⟩ discount; must equal the wallet signing the payment (x402 rail)","type":"string"},"prompt":{ ⟨3 unchanged words⟩ to render — call enhance_prompt first for bestresultsresults. Not needed for mode=tool models (upscale, remove-bg) — the image is the request","type":"string"},"resolution":{"description":"Video resolution tier,(videoor upscale tier for tool models)","type":"string"}},"required":["prompt"],"type":"object"}
{"destructiveHint":false,"idempotentHint":false,"openWorldHint":truefalse,"readOnlyHint":false}
⟨21 unchanged words⟩ {"description":"Text model id (default:deepseek/deepseek-v4-flashdeepseek/deepseek-v4.1-flash)","type":"string"},"payer": ⟨33 unchanged words⟩
⟨78 unchanged words⟩ at https://livepairai.com/auth/sign-up, create a key at/dashboard/settings/settings, send `x-api-key: lp_…`; the call bills your ⟨24 unchanged words⟩
Private LLM text completion (paid, x402) — Get the paid endpoint for OpenAI-compatible chat completions — frontier and open-weight models (DeepSeek, GPT, GLM, Qwen, Gemini, Llama, Kimi, Grok, Claude; list under textModels in GET /v1/agent/models). PAID call, two rails: x402 (endpoint answers 402, wallet settles, retry with PAYMENT-SIGNATURE → {choices, usage}) or account API key (`x-api-key: lp_…` — create at https://livepairai.com/dashboard/settings after signing up; bills prepaid credits, no 402). Price = estimated input tokens × input rate + max_tokens × output rate — set max_tokens tight to bill tight. Metering keeps token counts …
⟨53 unchanged words⟩ fastest t2i)","type":"string"},"payer":{"description":"Your 0x… wallet — applies your volume-tier discount; must equal the wallet signing the payment","type":"string"},"prompt":{"description":"The prompt to render ⟨22 unchanged words⟩
ListLivePair'sthe public image and video generationmodelscatalog — id, label, mode (text-to-image, image-to-video, etc),andrough generationtime.time, and private flag (private AI models — i2i/i2v source images must be AI-generated or owned, no real-person photos without consent).
⟨28 unchanged words⟩ from list_models","type":"string"},"payer":{"description":"Your 0x… payer wallet — applies your volume-tier discount","type":"string"},"resolution":{"description":"Video resolution tier", ⟨6 unchanged words⟩
Decide whichLivePairmodel fits a need — returns the modelId ⟨11 unchanged words⟩ or video; with or without a source image. Pass private:true to restrict to the private AI model line (i2v models).
⟨37 unchanged words⟩ image","video"],"type":"string"},"private":{"description":"True to consider only the private AI model line","type":"boolean"}},"required":["goal","media"],"type":"object"}
Keyword search overLivePair'sthe community prompt library — tested AI image and ⟨31 unchanged words⟩
Quote a generation price — Exact USDC price for a model+settings combo before paying — images are flat per call, video is per-second × duration at the resolution tier. The x402 endpoint always re-quotes from the body, so this is the same number it would charge.
Pick a model for a task — Decide which LivePair model fits a need — returns the modelId to use with generate_image. Cheapest, fastest, or highest quality; for images or video; with or without a source image.